Modeling and simulation of an industrial slurry-phase catalytic olefin polymerization reactor series

In the present study, a comprehensive mathematical model is developed to simulate the dynamic behaviour of an industrial slurry-phase olefin catalytic polymerization loop reactor series. More specifically, the effects of various operating conditions on the dynamic reactor behaviour (i.e., reaction temperature and pressure, inflow rates of catalyst, monomers and diluent, etc.) as well as the on the molecular and rheological polyolefin properties (i.e., M_n, M_w, MWD, complex viscosity, etc.) are fully assessed. According to the proposed modeling approach, each loop reactor (i.e., consisting of the loop reactor and the settling legs) is modeled as an ideal CSTR in series with a semi-continuous product removal unit. Dynamic macroscopic mass species and energy balances are derived to calculate the dynamic evolution of the concentrations of the various molecular species as well as of temperature profiles and heat removal in the two loop reactors. The polymer molecular properties (i.e., number- and weight-average molecular weights and molecular weight distribution) are determined by employing a generalized multi-site, Ziegler–Natta (Z–N) kinetic scheme in conjunction with the well-known method of moments. All the thermodynamic calculations, regarding the equilibrium species concentration in the various phases (i.e., solids and liquid), are carried out using the Sanchez–Lacombe Equation of State (S–L EOS). It is shown that the proposed comprehensive model is capable of simulating the dynamic operation of an industrial slurry-phase cascade-loop reactor series under different plant operating policies (i.e., start-up, grade transition, etc.).     

Parlay‐based service provision in circuit‐ and packet‐switched telecommunications networks

The spectrum of potential value added services over Internet telephony is wide, but the current service provision solutions are inadequate or proprietary. The nature of Internet differs significantly from that of circuit switched network, however, VoIP architectures can capitalize service control architectures in the PSTN world. We describe such an architecture based on the intelligent network and the Parlay, employing distributed objects and mobile agents as enabling technologies. This architecture has been implemented in the PSTN and the Internet and it has provided a framework for service provisioning, augmenting the space of supported services. Copyright © 2004 John Wiley & Sons, Ltd.     

Dynamic modeling of multizone,multifeed high-pressure LDPE autoclaves

A comprehensive mathematical model is developed to simulate the dynamic behavior of multizone, multifeed high-pressure ethylene polymerization autoclaves. To describe the complex flow patterns occurring in low-density polyethylene (LDPE) autoclaves, a user-specified multisegment, multirecycle model representation of the actual multizone reactor is established. A general reaction mechanism is employed to represent the kinetics of ethylene polymerization. Dynamic mass, molar species, and energy balances are derived to predict the polymerization rate, monomer conversion, molecular weight developments (e.g., M_n, M_w, long- and short-chain branching), and temperature profile with respect to time and spatial position in the reactor. Detailed results on the start-up and grade transition of a four-zone autoclave reactor are presented and the effects of the macromixing parameters (e.g., number of segments per reaction zone and the total and side external recycle ratios) on the dynamic behavior of the reactor are investigated. It is shown that the model macromixing parameters can significantly affect the initiator consumption rate in a reaction zone. The present model is capable of predicting accurately the dynamic behavior of LDPE autoclaves and, thus, can be employed in the design, optimization, and control of these reactors. © 1999 John Wiley & Sons, Inc. J Appl Polym Sci 73: 2327–2348, 1999     

Prediction of the molecular and polymer solution properties of LDPE in a high-pressure tubular reactor using a novel Monte Carlo approach

In the present work, a novel kinetic/topology Monte Carlo algorithm is developed for the prediction of molecular, topological and solution properties of highly branched low-density polyethylene (LDPE), produced in a high-pressure multi-zonal tubular reactor. It is shown that the combined kinetic/topology MC algorithm can provide comprehensive information regarding the distributed molecular and topological properties of LDPE (i.e., molecular weight distribution, short- and long-chain branching distributions, joint molecular weight-long chain branching distribution, branching order distribution, seniority/priority distributions, etc.) The molecular/topological results obtained from the MC algorithm are then introduced into a random-walk molecular simulator to calculate the solution properties of LDPE (i.e., the mean radius of gyration, R_g, and the branching factor, g) in terms of the chain length of the branched polyethylene. The validity of the commonly applied approximation regarding the random scission of highly branched polymer chains is assessed by a direct comparison of the average molecular properties of LDPE (i.e., number and weight average molecular weights), calculated by the combined kinetic/topology MC algorithm, with the respective predictions obtained by the commonly applied method of moments (MOM). Through this comparison it is demonstrated that the ambiguous implementation of the random scission reaction in the MOM formulation can result in erroneous predictions of the weight average molecular weight and MWD of LDPE. Finally, the effects of two key process parameters, namely, the polymerization temperature profile and the solvent concentration, on the molecular, topological and polymer solution properties of LDPE produced in a multi-zonal tubular reactor are investigated.     

Crawl and crowd to bring machine translation to under-resourced languages

We present a widely applicable methodology to bring machine translation (MT) to under-resourced languages in a cost-effective and rapid manner. Our proposal relies on web crawling to automatically acquire parallel data to train statistical MT systems if any such data can be found for the language pair and domain of interest. If that is not the case, we resort to (1) crowdsourcing to translate small amounts of text (hundreds of sentences), which are then used to tune statistical MT models, and (2) web crawling of vast amounts of monolingual data (millions of sentences), which are then used to build language models for MT. We apply these to two respective use-cases for Croatian, an under-resourced language that has gained relevance since it recently attained official status in the European Union. The first use-case regards tourism, given the importance of this sector to Croatia’s economy, while the second has to do with tweets, due to the growing importance of social media. For tourism, we crawl parallel data from 20 web domains using two state-of-the-art crawlers and explore how to combine the crawled data with bigger amounts of general-domain data. Our domain-adapted system is evaluated on a set of three additional tourism web domains and it outperforms the baseline in terms of automatic metrics and/or vocabulary coverage. In the social media use-case, we deal with tweets from the 2014 edition of the soccer World Cup. We build domain-adapted systems by (1) translating small amounts of tweets to be used for tuning by means of crowdsourcing and (2) crawling vast amounts of monolingual tweets. These systems outperform the baseline (Microsoft Bing) by 7.94 BLEU points (5.11 TER) for Croatian-to-English and by 2.17 points (1.94 TER) for English-to-Croatian on a test set translated by means of crowdsourcing. A complementary manual analysis sheds further light on these results.